Pergamon Tetrahedron Letters, Vo1. 37, No. 38, pp. 6831-6834, 1996

Copyright © 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved

PII: S0040-4039(96)01545-6 0040-4039196 $15.00 + 0.00

A Convenient Synthesis of Mono-6-hydroxy Permethylated I~-Cyclodextrin via tert.

Butyidimethylsilylation

Zhen Chen, Jerald S. Bradshaw,* and Milton L. Lee

Department of Chemistry and Biochemistry, Brigham Young University

Provo, LIT 84602 U.S.A.

Abstract: A convenient method to prepare mono-6-hydroxy permethylated I~--cyclodextrin (3) via tert- butyldimethylsilylation is described. Mono-6-tert-butylrYmaethylsilyl-~-eyclodextrin was prepared and treated with Nail and CH31 in the same pot. The protecting group was removed by refluxing with NI-~F in CH3OH to give 3 in an overall yield of 43%. Copyright © 1996 Elsevier Science Ltd

Recently, chemical modification of cyclodextrins with various functional groups has been extensively

investigated. Cyclodextrins are very popular building blocks for supramolecular structures1,2 and their

derivatives are widely used in the fields of analytical chemistry3 and as enzyme mimics. 4 However, selective

modifications of the cyclodextrins are difficult to control because of problems arising from steric and statistical

factors imposed by the torus structure and the large number of hydroxy groups. When a mono-functionalized

cyclodextrin is desired, the ideal situation is to have only one reactive functional group in the cyclodextrin

molecule with all other hydroxy groups blocked. In this case, positional isomers and homologous derivatives

with a higher degree of substitution can be avoided. The mono-substituted product can easily be separated.

Mono-6-hydroxy permethylated I~-cyclodextrin is such a compound which has been used in the synthesis of

photo- and electroactive receptors,5 stationary phases for liquid chromatography, 6 stationary phases for gas and

supercritical fluid chromatography,7 and cyclodextrin dimers.8 Unfortunately, the application of mono-6-

hydroxy permethylated 13-cyclodextrin is hindered by its inconvenient preparation. The generally used method

for its preparation was reported by Tanaka et al,6 although the experimental details and characterization of the

product were not given. First, mono-(6-O-trityl)-I~-cyclodextrin was prepared in about 30% yield by treating ~-

cyclodextrin with trityl chloride in pyridine.9 Second, all the remaining hydroxy groups were methylated in

dimethylformamide (DMF). Finally, the trityl group was removed in a two-phase system (concentrated HC1-

CHCI3). There are several disadvantages related to this synthetic method. The overall yield is not good (< 30%);

one must use pyridine in the first step and the work up is very tedious; and acid is used in the last step. A

convenient method to prepare mono-6-hydroxy permethylated 13-cyclodextrin in a reasonable yield would be

important for many workers in the field.

6831

6832

~ CH2OH)7 t-Bu(CHa)2SiCI Nail NH4F ~ (CH2OCHa)7 m- +

(OH)14 DMF, imidazole CH3~ CHa& v-~.(OCH3)14 1 2 (19%)

CH2OH

3 (43*/.)

Scheme 1. Synthesis of mono-6-hydroxy permethylated ~-cyclodextrin (3).

We now describe a convenient synthesis of the rifle compound 3 via tert-butyldimethylsilylation. The

strategy is shown in Scheme 1. The tert-butyldimethylsilyl group has proven to be a valuable protecting group

in cyclodextrin chemistry.10A] It can be selectively attached to the primary 6-OH groups of the cyclodextrin

glucose residues, and the silyl groups are stable under ordinary conditions. The silyl groups are easily removed

using NH4F, 12 avoiding the use of acid which can decompose the cyclodextdn. To a stirred mixture of 13-

cyclodextrin (4.54 g, 4 mmol) and imidazole (0.61 g, 6 mmol) in 100 ml of dry DMF, a solution of tert-

butyldimethylsilyl chloride (1.24 g, 8 mmol) in 20 ml of dry DMF was added dropwise at rt under an Argon

atmosphere. The mixture was stirred at rt while carefully monitoring the reaction by TLC on silica gel (1-

BuOH/EtOH/H20:5/4/3 by volume). The reaction reached equilibrium and no further apparent change occurred

after 1 h. TLC showed three major components having Rf values of 0.41 ([3-cyclodextrin), 0.63 and 0.72. After

cooling to 0 oc, Nail (6.4 g, 260 mmol) was added and the mixture was stirred at 0 o(2 for 30 rain then at rt for

1 h. After cooling to 0 oc, 34 ml of CH3I (77 g, 545 retool) was added dropwis¢. The mixture was kept for I h

at.0 oC then for 24 h at rt. TLC (CHCI3/CH3OH: 10/1) showed three components with~Rf values of 0.68

(permethylated ~-cyclodextdn, by comparing with the standard sample), 0.72 and 0.78. Because there was no

component having an Rf value smaller than that of permethylated ~-cyclodextrin, all hydroxy groups must be

methylated. Although the migration of tert-butyldimethylsilyl groups from O-2 to 0-3 during the alkylation

process is usual, the same groups at the 6 positions are stable. 13 The reaction mixture was treated using a

procedure similar to that reported.14 Excess Nail was decomposed by the addition of CH3OH (15 ml) after

cooling the system to 0 oC. The resulting mixture was poured into 400 ml of ice water with stirring and

extracted with CHC13 ( 4 x 100 ml). The combined CHC13 layers were washed successively with 75 ml of a 3%

aqueous solution of Na2S203 to remove excess CH3I and H20 (3 x 75 ml) and then dried (MgSO4). After

concentrating and drying under vacuum overnight at 80 oc, a solid residue of 6.11 g was obtained. Trying to

separate the three components at this stage by the usual column chromatography was very difficult because their

polarities are very similar. The mixture was subjected to the deprotcction step before separation was achieved.

The residue (6.11 g) was refluxed with 2.06 g of NH4F in 200 ml of CH3OH for 24 h and then

concentrated.12 The solution of the residue in ethyl acetate was filtered through celite, and the filtrate was

concentrated. TLC (CHC13/CH3OH: 10/1) revealed three components with Rf values of 0.68, 0.54, and 0.42,

respectively. Column chromatography (CHC13/CH3OH: 50/1) on silica gel easily gave pure tmrmethylated ~-

6833

cyclodexn-in (2) 1.08 g (19%, Rf = 0.68) and mono-6-O-hydroxy pennethylated l~yclodcx~_n (3) 2.43 g (43%,

Rf = 0.54). When using one equivalent of tert-butyldimethylsilyl chloride in the first step, only 23% of 3 was

isolated. The third spot, which is a mixture of isomeric dihydroxy compounds, was not isolated. Product 3 thus

obtained was fully c h ~ by IH and t3C NMR spectroscopy, FABMS and elemental analysis.15

A

B

(~-12

L__ C~3

~ 4 s 1 "0"

fl C-5, C-6 2-0CH3

3-OCHs

n ~ . = l , , , . p , , , r . , r . , t . . , , i , . . . i , , , , i , , , , i , . , r . , i , . , , i . I , v . . . t , , . , I i , . . p , , , i , , , , p , , , p . , , i . , , , t , , , , p . , , t , , , ' l " ' - l ' " ' l ' " ; t ' " ; t " ' " , Z , l , , , ~ ' l S , - i , , , . i , , , . i . , , , I , , . , i , . , , i - , . i - , , i , , , , i , , , ' l

71.5 71 .1 70. '7 ppm 61 .7 61 .3 6 0 . 9 ppm 5 8 . 8 58 .5 5 8 . 2 ppm

Figure 1. Partial DEPT (A) and 13C NMR 03) spectra of 3.

Mono-6-hydroxy permethylated ~-cyclodextrin (3) is an unsymmetric molecule. Figure I shows partial

DEPT (A) and 13C NMR (B) spectra of 3. The DEFT specmun is unique in that peaks corresponding to methyl

and methine carbon atoms point up and those for methylene carbon atoms point down. 10 In general, the signal

for the C-6 atom containing an OH group appears at about 60 ppm. O-Methylation of the hydroxy group is

expected to shift the a-carbon m about 70 ppm.lS In the DEPT spectrum of 3, the signals for the six CI'h carbon

atoms containing OCH3 groups at the 6-position appear at 70.8-71.3 ppm and there am six of these peaks when

one accounts for the fact that the signals for one CH2 and one CH appear at the same ppm. The remaining CH2

carbon atom produces a ~ignnl at 61.3 ppm indicating a CHzOH. The 13C NMR spcctnma shows seven signals

each for the 2- and 3-OCH 3 groups hut only six signals for the 6-OCH3 groups. These spectral data clearty show

that 3 is indeed the mono-6-hydroxy permethylated compound.

In conclusion, the details of the synthesis and characterization of mono-6-hydroxy permethylated

6834

cyclodextrin (3) are reported. The method describea:l here is superior to the earlier one.6 The overall yield is

higher, the first two steps are in one pot, and the final work up is simple. Compound 3 is not only an appropriate

nucleophile, but it also can be transformed easily into the mono-tosylate, aldehyde, carboxylic acid and other

mono derivatives on the primary side of the 13-cyclodextrin. Thus, a wide application of compound 3 can be

envisioned.

Acknowledgement: The authors thank Dr. Yoichi Habata for many helpful discussions, and Dr. Du Li

for obtaining the high resolution NMR spectra.

REFERENCES AND NOTES

1. Wenz, G. Angew. Chem. Int. Ed. Engl. 1994, 33, 803.

2. Amabilino, D. B.; Stoddart, J. F. Chem. Rev. 1995, 95, 2725.

3. Li, S.; Purdy, W. C. Chem. Rev. 1992, 92, 1457.

4. Tabushi, I: Functionalized cyclodextrins as artificial enzymes. In Chemical approaches to

understanding enzyme catalysis; Green, B.; Ashani, Y.; Chipman, D. Eds.; Elevier, Amsterdam. 1982.

5. Brugger, N.; Deschenaux, R.; Ruch, T.; Ziessel, R. Tetrahedron Lett. 1992, 33, 3871.

6. Tanaka, M.; Kawaguchi, Y.; Niinae, T.; Shono, T. J. Chromatogr. 1984, 314, 193.

7. Yi, G-L.; Bradshaw, J. S.; Rossiter, B. E.; Malik, A.; Li, W.-B.; Lee, M. L. J.Org.Chem. 1993, 58,

4845.

8. Deschenaux, R.; Greppi, A.; Ruth, T.; Kriemler, H-P.; Raschdorf, F.; Ziessel, R. TetrahedronLett.

1994, 35, 2165.

9. Melton, L. D.; Slessor, K. N. Carbohydr. Res. 1971, 18, 29.

10. Pregel, M. J.; Buncel, E. Can. J. Chem. 1991, 69, 130.

11. Takeo, K.; Mitoh, H.; Uemura, K. Carbohydr. Res. 1989, 187, 203.

12. Chang, W-J,; Robins, M. J. Tetrahedron Lett. 1992, 33, 1177.

13. Icheln, D.; Gehrcke, B.; Piprek, Y.; Mischnick, P.; Konig, W.; Dessoy, M. A.; Morel, A. F. Carbohydr.

Res. 1996, 280, 237.

14. Szejfli, J.; Liptak, A.; Jodal, I.; Fuged, P.; Nanasi, P.; Neszmelyi, A. Starch 1980, 32, 165.

15. Selected analytical data for 3:

IH NMR (CDCI3) 8 5.2-4.9 (m, 7 H), 3.9-3.0 (m, 103 H), of which 3.34 (s), 3.45 (s) and 3.56 (s) are for

three OCH3 groups; 13C NMR (CDC13) ~ 98.04, 98.00, 97.98, 97.94, 97.92, 97.82, 81.41, 81.19, 81.10,

81.04, 81.00, 80.93, 80.90, 80.83, 80.81, 80.76, 80.69, 80.55, 80.08, 79.66, 79.49, 79.07, 78.86, 77.67,

76.52, 76.46, 76.26, 76.00, 70.83, 70.70, 70.61, 70.51, 70.44, 70.34, 70.27, 70.19, 70.07, 69.89, 60.67,

60.59, 60.57, 60.55, 60.48, 60.32, 60.27, 60.06, 58.12, 58.08, 57.97, 57.93, 57.68, 57.64, 57.47, 57.35,

57.31, 57.26; FABMS: m/z 1437, [M-H+Na]+; Anal. Calcd for C62Hl10035: C, 52.61; H, 7.83. Found:

C, 52.40; H, 7.67. 16. Boger, J.; Corcoran, R. J.; Lehn, J-M. Helv. Chim. Acta 1978, 61, 2190.

(Received in USA 18 June 1996; revised 29 July 1996; accepted 1 August 1996)